Taking the enterprise of the future to the next level
When we began our journey in search of the enterprise of the future eight years ago, one of the first things we did was build the four-pillar framework, which has pretty much held to this day. The four pillars are: new business ecosystems and strategies (leadership); new organizational designs (organization); new living, working and learning environments (learning); and technology infrastructure nexus (technology).
You can review the full list of attributes for each pillar in the October 2006 issue of KMWorld magazine: kmworld.com/Articles/Column/The-Future-of-the-Future/The-Future-of-the-Future-Building-the-enterprise-of-the-future-a-framework-for-trans
formation-18290.aspx. Let's take a look at a few of those attributes, and see how close we've come to making them real. But first, a word about predictions.
In the world of quantum physics, you can only know with high accuracy the position of a particle or its momentum, but not both. Predicting the future is much the same. An event can have a high probability of occurring (position); you just don't know precisely when (momentum). Just ask any stock trader who has attempted to ride one of the many market "bubbles" we've seen.
Similarly, you can know precisely when an event such as an election, or a book or movie release will occur. You just don't know the outcome, at least not with high precision.
With those caveats in mind, let's take a look at four trends that will directly impact the future of the future. We will link each trend to the pillar impacted the most. In deference to Werner Heisenberg and his uncertainty principle, we will look mostly at position (where we are with respect to the development of a particular trend), while sidestepping any temptation to predict when the full impact of that trend will come into play.
Trend 1: anticipatory systems
Attribute— anticipatory business intelligence (BI) systems
One of our biggest surprises was how little anticipatory systems have moved into practice. Although business intelligence is currently a hot topic, most of the work in that area has been focused on looking in the rearview mirror and extrapolating forward.
Thirty years ago, Robert Rosen, a brilliant biologist and mathematician, postulated a theory based on how living systems anticipate-a trait necessary for survival. His thesis statement, considered radical at the time, was: The current state of an anticipatory system, including living organisms, is determined by a predicted future state.
Another jewel in Rosen's work is his theory of categories, which postulates how latent categorization errors propagate through an event chain, creating major problems downstream. All of that is a fertile field for KM'ers, and presents a huge opportunity to add value and overcome many of the shortfalls of BI. Look for a growing emphasis on the development of anticipatory models to dramatically increase the effectiveness of BI.
Trend 2: sharable ontologies
Attribute—people actively participating in cross-boundary learning groups relevant to their expertise, and practicing open communications and the free flow of knowledge across the enterprise.
In previous articles, we've mentioned the many challenges associated with collaborative knowledge sharing, especially bridging differences in perspectives, semantics, context and other factors. Anyone who has attempted to build a shared knowledge repository knows firsthand the investment that needs to be made in organizing and curating knowledge spaces in ways that can be easily searched, augmented (e.g., through inference) and shared across teams, organizational units and disciplines.
After a flurry of activity sponsored by DARPA during the 1990s to early 2000s, the momentum in this area has slowed considerably. Standards painstakingly developed during this period are there for the taking: web ontology language (OWL), resource description framework (RDF), and ontology inference layer (OIL), to name a few. Their scope, however, was limited for two reasons. First, expectations were carefully managed by not trying to solve all of the challenges associated with knowledge representation at once. Second, as is the case for most software-related standards, they are by necessity low-ambiguity languages.
That creates a huge mismatch. Low-ambiguity languages will never be able to support knowledge exchange at the level that is needed to perform in a complex, rapidly changing world. That may be the biggest obstacle preventing the semantic web from coming to full fruition. We need to expand the ontology "stack" to higher levels, moving beyond the database-driven notion that everything can be expressed using the basic <subject, predicate, object> schema.
Trend 3: deep learning
Attribute—workspaces integrating physical, information, cognitive and organizational needs.
Of those four, the least progress has been made in the cognitive area, but that is starting to change. Pent-up frustration associated with knowledge transfer, especially in response to the wave of retirees, is driving change. While Trend 2 continues to expand the capability to capture and transfer explicit knowledge, the age-old problem of tacit knowledge just won't go away.
In education, science, politics and most of the many other areas of human engagement, our sound bite mentality is actually creating a learning disability, in which students and mentees are losing the capacity for critical thinking, creative problem-solving and adaptive learning through self-discovery. Whether through short stories or long-winded lectures, knowledge transfer is becoming dangerously shallow.
Deep learning, which has been around since the time of Socrates, occurs at the memory engram level. Karl Pribham and his colleagues have produced a treasure trove of brain research in that area. As the research moves into practice, look for real breakthroughs in knowledge transfer as mentors, trainers, educators, even knowledge engineers, learn to apply deep learning models and methods.